Income-based deductibles (IBDs) for prescription drugs require elderly patients to pay all drug costs up to an amount determined by their incomes; this approach has been proposed for containing costs in a potential Medicare drug benefit program. Older patients taking antidepressants may be especially vulnerable to adverse effects from IBDs because of their limited resources, frequent need for multiple drugs, susceptibility to inappropriately taken regimens, and stigma associated with taking antidepressants. On the other hand, antidepressants are the single largest drug expenditure in many health systems. It is possible that IBDs may contain antidepressant costs without adverse effects, ensuring the fiscal viability of drug assistance programs. In July 2003, British Columbia residents equal to or >65 (N equal to approximately 520,000) will begin an IBD program, paying all drug costs up to 2-4% of their income out-of-pocket. We will use comprehensive drug and health care utilization databases and patient-centered information from interviews to evaluate this substantial coverage change in a large, stable population of elderly antidepressant users. Interrupted time-series regression analyses will identify the impact of IBDs on completed suicides and attempts, hospitalizations, emergency room care, nursing home stays, outpatient visits, antidepressant discontinuation, use of suboptimal regimens, overall net monetary savings, and out-of-pocket expenses for antidepressants. Elderly who discontinue or change to potentially suboptimal regimens are most likely to suffer adverse outcomes from IBDs, yet even severe effects in them may be missed in overall analyses. We will focus on high-risk patients who discontinue or change regimens to see if IBDs negatively impact health, utilization, or economic outcomes. Using patient-centered information captured in interviews, we will identify the burdens and coping strategies elderly use to deal with greater antidepressant cost-sharing, such as reducing drug use, obtaining nontraditional sources of drugs, or doing without necessities. We will examine whether depression severity and other burdens are greater during annual periods when patients are fully responsible for paying vs. fully reimbursed for their drugs. We will conduct extensive dissemination activities, including convening an international workshop on drug cost containment policies among elderly users of psychiatric medications. We will also use other national and international forums to inform ongoing debates over drug coverage for vulnerable elderly patients with depression.